REVERSE model: a novel approach in medical research.

Area of Science:

• Medical research methodology
• Data science in healthcare
• Biostatistics

Background:

• Randomized controlled trials (RCTs) are the gold standard but are costly and have limited external validity.
• The REVERSE model is proposed to overcome these limitations.
• Existing datasets can be leveraged to address new research questions efficiently.

Purpose of the Study:

• To introduce and validate the REVERSE model for repurposing existing datasets.
• To demonstrate the model's ability to mitigate biases inherent in observational data.
• To replicate findings from previous clinical trials using the REVERSE model.

Main Methods:

• The REVERSE model involves two phases: data mining and randomization.
• Data mining merges compatible datasets using strict inclusion/exclusion criteria to reduce selection bias.
• The randomization phase addresses dataset biases by organizing data into sub-cohorts and testing new factors in concordant groups.

Main Results:

• The REVERSE model accurately replicated findings from the ROMICAT-II and ROMICAT-I trials using only ROMICAT-II data.
• For ROMICAT-II, hospitalization stay differences were matched 100% with the original trial.
• For ROMICAT-I, plaque prevalence at cardiac computed tomography angiography (CCTA) was accurately reproduced.

Conclusions:

• The REVERSE model enables the repurposing of existing datasets for novel research.
• It effectively mitigates inherent biases through stringent data matching and randomization.
• This approach offers a cost-effective and valid alternative to traditional RCTs for certain research questions.